This proposal will test the hypothesis that the occurrence of hypocomplementemia, proteinuria, thrombocytopenia, and lupus anticoagulant during the course of pregnancy in patients with systemic lupus erythematosus (SLE) does not represent exacerbation of immunologic SLE, and that by implication hypocomplementemia in this setting is likely due to reduced synthesis rather than increased degradation of complement. A new assay, the measurement of C1s-C1r-C1 inhibitor complex (C1s-C1Inh complex), is used as a quantitative measure of complement consumption. With this assay, in preliminary studies, hypocomplementemic pregnant SLE patients with proteinuria have normal values, while hypocomplementemic not pregnant SLE nephritis patients with proteinuria have very abnormal values. Hypocomplementemia (CH50, C3 and C4) is linearly related to C1s-C1Inh complex in not pregnant patients but there is no correlation between complement and C1s-C1Inh complex in pregnant SLE patients. Approximately 60 pregnant SLE patients will undergo serial serological and clinical evaluations through 6 months post-partum to evaluate "flare" status. Clinical and serological comparisons of pregnant SLE patients will be with SLE not pregnant patients. Studies of antibody to SSA (associated with congenital SLE), and placental and ovarian hormone function will attempt to predict fetal survival. The hormone assays will compare SLE pregnancies with diabetes pregnancies. We predict that pregnant (hypocomplementemic) SLE patients with proteinuria, thrombocytopenia and lupus anticoagulant will have normal C1s-C1Inh complex, and thus will contrast with not pregnant SLE patients. We also predict that fetal survival will be a function of measureable abnormalities in placental or ovarian hormone and/or SSA antibody.